The challenge for Day 2 of dashboard week was to use data from this Star Wars API. The API has data on everything from planets to characters and even vehicles. We were told we could do anything we wanted with the data so I decided to create a dashboard to help find the best vehicle in the Star Wars universe.

Data extraction and preparation:
Alteryx was used to extract the data from the API. First I used a Text Input Tool which contained the relevant URLs then a Download Tool to download the data and a JSON Parse tool to parse it. Once I had the data I needed to transform and clean it using Filter, Text To Column, Cross-Tab, Formula and Data Cleansing tools.

Dashboard:
I ended up using 3 tables, each containing data about the vehicles, characters and films, I simply created relationships in Tableau between each of these and started building by dashboard.

My first worksheet categorises each vehicle into a class and displays its picture. I already had a “class” column, however I found there were too many obscure classes so I created a group and grouped as many of the classes together as I could. I then put this new class group on rows, changed the mark to “shape” and the “vehicle name” column on the shapes card. I then assigned each vehicle name a custom shape of its picture and sorted each class group by count distinct of vehicle name.

Next I wanted to compare the relationship between measures (such as speed, capacity, cost etc.) for all of the vehicles so I created a scatter plot. I wanted the user to have the ability to choose which measures were most important to them so I created 2 parameters and 2 calculated fields (see below images), I placed each of the calculated fields on columns and rows and “vehicle names” on the detail mark and voila. I also created a simple bar chart with the vehicle names on rows and one of the calculated fields on columns and sorted in descending order so that the user can have a clear ranking of each vehicle by the measure they select with the parameter.
       

My final worksheet was a donut chart with the sum of the calculated field I created, I placed the class group on the colour card and the calculated field on the label card and performed a quick table calc to calculate the % of total for each vehicle class for the selected measure. Read my blog here for an in depth guide on how to create a donut chart.

I then finished with a list of films and characters, which I applied custom shares to, so once an interactive filter is applied to the dashboard the characters who have piloted a selected vehicle and the films in which the vehicles appear are displayed. I also applied a list of filters for each measure across the dashboard and displayed the 2 parameters so the user can select which measures they would like to compare. Here is the final product:

Insights:

  • There is a positive relationship between passengers, crew, cargo capacity, length and cost measures.
  • Surprisingly there is no real relationship between cost and max speed
  • The second fastest vehicle, the T-16 Skyhopper is only 14.5K credits which is well below the 32.9K avg.

 

 

The Data School
Author: The Data School